![]() Molecular structure of the PDZ domain included in the humanGOPC (Golgi-associated PDZ and coiled-coil motif-containing protein) protein | |||||||||||
Identifiers | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Symbol | PDZ | ||||||||||
Pfam | PF00595 | ||||||||||
InterPro | IPR001478 | ||||||||||
SMART | PDZ | ||||||||||
PROSITE | PDOC50106 | ||||||||||
SCOP2 | 1lcy /SCOPe /SUPFAM | ||||||||||
CDD | cd00136 | ||||||||||
|
ThePDZ domain is a commonstructural domain of 80-90amino-acids found in thesignalingproteins ofbacteria,yeast,plants,viruses[1] andanimals.[2] Proteins containing PDZ domains play a key role in anchoring receptor proteins in the membrane to cytoskeletal components. Proteins with these domains help hold together and organize signaling complexes at cellular membranes. These domains play a key role in the formation and function of signal transduction complexes.[3] PDZ domains also play a highly significant role in the anchoring ofcell surface receptors (such asCftr andFZD7) to theactincytoskeleton via mediators likeNHERF andezrin.[4]
PDZ is aninitialism combining the first letters of the first three proteins discovered to share the domain —post synaptic density protein (PSD95),Drosophila disc large tumor suppressor (Dlg1), andzonula occludens-1 protein (zo-1).[5] PDZ domains have previously been referred to as DHR (Dlg homologous region)[6] or GLGF (glycine-leucine-glycine-phenylalanine) domains.[7]
In general PDZ domains bind to a short region of theC-terminus of other specific proteins. These short regions bind to the PDZ domain bybeta sheet augmentation. This means that the beta sheet in the PDZ domain is extended by the addition of a further beta strand from the tail of the binding partner protein.[8] The C-terminal carboxylate group is bound by anest (protein structural motif) in the PDZ domain, i.e. aPDZ-binding motif.
PDZ is an acronym derived from the names of the first proteins in which the domain was observed. Post-synaptic density protein 95 (PSD-95) is asynaptic protein found only in the brain.[7] Drosophila disc large tumor suppressor (Dlg1) and zona occludens 1 (ZO-1) both play an important role atcell junctions and in cell signaling complexes.[9] Since the discovery of PDZ domains more than 20 years ago, hundreds of additional PDZ domains have been identified. The first published use of the phrase “PDZ domain” was not in a paper, but a letter. In September 1995, Dr.Mary B. Kennedy of theCalifornia Institute of Technology wrote a letter of correction to Trends in Biomedical Sciences.[10] Earlier that year, another set of scientists had claimed to discover a new protein domain which they called a DHR domain.[6] Dr. Kennedy refuted that her lab had previously described exactly the same domain as a series of “GLGF repeats”.[7] She continued to explain that in order to “better reflect the origin and distribution of the domain,” the new title of the domain would be changed. Thus, the name “PDZ domain” was introduced to the world.
PDZ domain structure is partially conserved across the various proteins that contain them. They usually have 5-6β-strands and one short and one longα-helix. Apart from this conserved fold, thesecondary structure differs across PDZ domains.[3] This domain tends to be globular with a diameter of about 35 Å.[11]
When studied, PDZ domains are usually isolated asmonomers, however some PDZ proteins formdimers. The function of PDZ dimers as compared to monomers is not yet known.[3]
A commonly accepted theory for thebinding pocket of the PDZ domain is that it is constituted by severalhydrophobic amino acids, apart from the GLGF sequence mentioned earlier, the mainchain atoms of which form anest (protein structural motif) binding the C-terminal carboxylate of the protein or peptide ligand. Most PDZ domains have such a binding site located between one of the β-strands and the long α-helix.[12]
PDZ domains have two main functions: Localizing cellular elements, and regulating cellular pathways.
The first discovered function of the PDZ domains was to anchor receptor proteins in the membrane to cytoskeletal components. PDZ domains also have regulatory functions on different signaling pathways.[13] Any protein may have one or several PDZ domains, which can be identical or unique (see figure to right). This variety allows these proteins to be very versatile in their interactions. Different PDZ domains in the same protein can have different roles, each binding a different part of the target protein or a different protein altogether.[14]
PDZ domains play a vital role in organizing and maintaining complex scaffolding formations.
PDZ domains are found in diverse proteins, but all assist in localization of cellular elements. PDZ domains are primarily involved in anchoringreceptor proteins to thecytoskeleton. For cells to function properly it is important for components—proteins and other molecules— to be in the right place at the right time. Proteins with PDZ domains bind different components to ensure correct arrangements.[13] In theneuron, making sense ofneurotransmitter activity requires specific receptors to be located in thelipid membrane at the synapse. PDZ domains are crucial to this receptor localization process.[15] Proteins with PDZ domains generally associate with both the C-terminus of the receptor and cytoskeletal elements in order to anchor the receptor to the cytoskeleton and keep it in place.[14][16] Without such an interaction, receptors would diffuse out of the synapse due to the fluid nature of the lipid membrane.
PDZ domains are also utilized to localize elements other than receptor proteins. In the human brain,nitric oxide often acts in the synapse to modifycGMP levels in response toNMDA receptor activation.[17] In order to ensure a favorable spatial arrangements, neuronalnitric oxide synthase (nNOS) is brought close to NMDA receptors via interactions with PDZ domains on PSD-95, which concurrently binds nNOS andNMDA receptors.[16] With nNOS located closely to NMDA receptors, it will be activated immediately after calcium ions begin entering the cell.
PDZ domains are directly involved in the regulation of different cellular pathways. This mechanism of this regulation varies as PDZ domains are able to interact with a range of cellular components. This regulation is usually a result of the co-localization of multiple signaling molecules such as in the example with nNos and NMDA receptors.[16] Some examples of signaling pathway regulation executed by the PDZ domain include phosphataseenzyme activity,mechanosensory signaling, and the sorting pathway ofendocytosed receptor proteins.
The signaling pathway of thehuman protein tyrosine phosphatase non-receptor type 4 (PTPN4) is regulated by PDZ domains. This protein is involved in regulatingcell death. Normally the PDZ domain of this enzyme is unbound. In this unbound state the enzyme is active and prevents cell signaling forapoptosis. Binding the PDZ domain of this phosphatase results in a loss of enzyme activity, which leads to apoptosis. The normal regulation of this enzyme prevents cells from prematurely going through apoptosis. When the regulation of the PTPN4 enzyme is lost, there is increasedoncogenic activity as the cells are able toproliferate.[18]
PDZ domains also have a regulatory role in mechanosensory signaling inproprioceptors andvestibular andauditoryhair cells. The proteinWhirlin (WHRN) localizes in thepost-synaptic neurons of hair cells that transform mechanical movement intoaction potentials that the body can interpret. WHRN proteins contains three PDZ domains. The domains located near theN-terminus bind to receptor proteins and other signaling components. When the one of these PDZ domains is inhibited, the signaling pathways of the neurons are disrupted, resulting in auditory, visual, and vestibular impairment. This regulation is thought to be based on the physical positioning WHRN and the selectivity of its PDZ domain.[19]
Regulation of receptor proteins occurs when the PDZ domain on theEBP50 protein binds to the C-terminus of thebeta-2 adrenergic receptor (β2-AR). EBP50 also associates with a complex that connects toactin, thus serving as a link between the cytoskeleton and β2-AR. The β2-AR receptor is eventually endocytosed, where it will either be consigned to alysosome for degradation or recycled back to the cell membrane. Scientists have demonstrated that when the Ser-411 residue of the β2-AR PDZ binding domain, which interacts directly with EBP50, is phosphorylated, the receptor is degraded. If Ser-411 is left unmodified, the receptor is recycled.[20] The role played by PDZ domains and their binding sites indicate a regulative relevance beyond simply receptor protein localization.
PDZ domains are being studied further to better understand the role they play in different signaling pathways. Research has increased as these domains have been linked to different diseases including cancer as discussed above.[21]
PDZ domain function can be both inhibited and activated by various mechanisms. Two of the most prevalent include allosteric interactions and posttranslational modifications.[3]
The most commonpost-traslational modification seen on PDZ domains isphosphorylation.[22] This modification is primarily aninhibitor of PDZ domain andligand activity. In some examples, phosphorylation of amino acid side chains eliminates the ability of the PDZ domain to formhydrogen bonds, disrupting the normal binding patterns. The end result is a loss of PDZ domain function and further signaling.[23] Another way phosphorylation can disrupt regular PDZ domain function is by altering thecharge ratio and further affecting binding and signaling.[24] In rare cases researchers have seen post-translational modifications activate PDZ domain activity[25] but these cases are few.
Another post-translational modification that can regulate PDZ domains is the formation ofdisulfide bridges. Many PDZ domains containcysteines and are susceptible to disulfide bond formation inoxidizing conditions. This modification acts primarily as an inhibitor of PDZ domain function.[26]
Protein-protein interactions have been observed to alter the effectiveness of PDZ domains binding to ligands. These studies show thatallosteric effects of certain proteins can affect the binding affinity for differentsubstrates. Different PDZ domains can even have this allosteric effect on each other. This PDZ-PDZ interaction only acts as an inhibitor.[27] Other experiments have shown that certainenzymes can enhance the binding of PDZ domains. Researchers found that the proteinezrin enhances the binding of the PDZ proteinNHERF1.[4]
PDZ proteins are a family of proteins that contain the PDZ domain. This sequence of amino-acids is found in many thousands of known proteins. PDZ domain proteins are widespread ineukaryotes andeubacteria,[2] whereas there are very few examples of the protein inarchaea. PDZ domains are often associated with otherprotein domains and these combinations allow them to carry out their specific functions. Three of the most well documented PDZ proteins arePSD-95,GRIP, andHOMER.
PSD-95 is a brain synaptic protein with three PDZ domains, each with unique properties and structures that allow PSD-95 to function in many ways. In general, the first two PDZ domains interact with receptors and the third interacts with cytoskeleton-related proteins. The main receptors associated with PSD-95 areNMDA receptors. The first two PDZ domains of PSD-95 bind to the C-terminus of NMDA receptors and anchor them in the membrane at the point of neurotransmitter release.[28] The first two PDZ domains can also interact in a similar fashion withShaker-type K+ channels.[28] A PDZ interaction between PSD-95,nNOS andsyntrophin is mediated by the second PDZ domain. The third and final PDZ domain links to cysteine-rich PDZ-binding protein (CRIPT), which allows PSD-95 to associate with thecytoskeleton.[28]
Glutamate receptor interacting protein (GRIP) is a post-synaptic protein that interacts withAMPA receptors in a fashion analogous to PSD-95 interactions with NMDA receptors. When researchers noticed apparent structuralhomology between the C-termini of AMPA receptors and NMDA receptors, they attempted to determine if a similar PDZ interaction was occurring.[29] Ayeast two-hybrid system helped them discover that out of GRIP's seven PDZ domains, two (domains four and five) were essential for binding of GRIP to the AMPA subunit called GluR2.[14] This interaction is vital for proper localization of AMPA receptors, which play a large part inmemory storage. Other researchers discovered that domains six and seven of GRIP are responsible for connecting GRIP to a family ofreceptor tyrosine kinases calledephrin receptors, which are important signaling proteins.[30] A clinical study concluded thatFraser syndrome, anautosomal recessive syndrome that can cause severe deformations, can be caused by a simple mutation in GRIP.[31]
HOMER differs significantly from many known PDZ proteins, including GRIP and PSD-95. Instead of mediating receptors near ion channels, as is the case with GRIP and PSD-95, HOMER is involved inmetabotropic glutamate signaling.[32] Another unique aspect of HOMER is that it only contains a single PDZ domain, which mediates interactions between HOMER and type 5 metabotropic glutamate receptor (mGluR5).[15] The single GLGF repeat on HOMER binds amino acids on the C-terminus of mGluR5. HOMER expression is measured at high levels during embryologic stages in rats, suggesting an important developmental function.[15]
There are roughly 260 PDZ domains in humans. Several proteins contain multiple PDZ domains, so the number of unique PDZ-containing proteins is closer to 180. In the table below are some of the better studied members of this family:
Studied PDZ Proteins | |||
---|---|---|---|
Erbin | GRIP | Htra1 | Htra2 |
Htra3 | PSD-95 | SAP97 | CARD10 |
CARD11 | CARD14 | PTP-BL[33] |
The table below contains all known PDZ proteins in humans (alphabetical):
There is currently one known virus containing PDZ domains:
Viruses | |
---|---|
Tax1 |